Bullish

DeepSeek V4 Pro API Launches With 3 Yuan Input Pricing and Enhanced Agent Capabilities

07:34

DeepSeek V4 Pro integrates into Responses API with 3 yuan input pricing. It outperforms rivals in CyberGym and AutomationBench benchmarks, signaling strong agent utility.

Woofun AI reports that the official DeepSeek V4 Pro model has been integrated into the Responses API, retaining its existing model name. The update supports integration with both the Responses API and Codex, significantly boosting agent capabilities.

API documentation sets input pricing at 3 yuan per million tokens and output at 6 yuan, with cached inputs at 0.025 yuan. In benchmark tests, V4 Pro scored 87.9 on Terminal Bench, nearing Fable 5's 88. It surpassed Fable 5 in the CyberGym security benchmark and the AutomationBench agent evaluation.

WOOFUN AI

Impact Assessment · Quick Read

The competitive pricing of DeepSeek V4 Pro lowers barriers for high-performance agent deployment, potentially accelerating adoption in enterprise workflows. Outperforming established models in security and automation benchmarks suggests a shift in the competitive landscape for specialized AI agents. This may pressure competitors to adjust pricing or enhance their own agent-specific capabilities.
Generated by WOOFUN AI · For reference only, not investment advice

Comments

Me
Replying to @User
0/800

No comments yet.

Notifications

Sign in to view messages
View all messagesManage subscriptions